Update science_news.recipe

Pull articles of latest magazine issue instead of RSS feed.

recipes/science_news.recipe

@@ -1,20 +1,18 @@
 #!/usr/bin/env python
 
 __license__ = 'GPL v3'
+
 '''
 sciencenews.org
 '''
 
 from calibre.web.feeds.news import BasicNewsRecipe
 import datetime
-
+import re
 
 class ScienceNewsIssue(BasicNewsRecipe):
     title = u'Science News'
-    __author__ = u'Darko Miletic, Sujata Raman and Starson17'
-    description = u'''Science News is an award-winning bi-weekly
-    newsmagazine covering the most important research in all fields of science.
-    This recipe downloads the last 2 weeks of articles.'''
+    description = u"Science News is an award-winning bi-weekly newsmagazine covering the most important research in all fields of science. This recipe downloads all the articles from the latest issue."
     category = u'Science, Technology, News'
     publisher = u'Society for Science & the Public'
     oldest_article = 14
@@ -43,15 +41,55 @@ class ScienceNewsIssue(BasicNewsRecipe):
         )
     ]
 
-    feeds = [(u"Articles", u'https://www.sciencenews.org/feed')]
+    def parse_index(self):
 
-    def get_cover_url(self):
-        d = datetime.date(2022, 3, 26)
+        # Get URL of latest mag page
+        ld = self._get_mag_date()
+        url = f"https://www.sciencenews.org/sn-magazine/{ld:%B}-{ld.day}-{ld.year}"
+        url = url.lower()
+
+        # Get articles
+        soup = self.index_to_soup(url)
+        re_article = re.compile("https://www.sciencenews.org/article/")
+        stories = []
+        past_urls = set()
+        for sec in soup.find_all(href=re_article):
+
+            article_url = sec["href"]
+            article_title = sec.text.strip()
+
+            # Ignore image URLs which do not have text title
+            if article_title == "":
+                continue
+
+            # Ignore if link is a duplicate
+            if article_url in past_urls:
+                continue
+
+            past_urls.add(article_url)
+            article_info = {
+                "url": article_url,
+                "title": article_title,
+            }
+            stories.append(article_info)
+
+        index = [("Articles", stories),]
+        return index
+
+    def _get_mag_date(self):
+        """Return date of latest magazine issue.
+        It is published every 2 weeks."""
+
+        d = datetime.date(2022, 6, 18)
         t = datetime.date.today()
         ld = None
         while d <= t:
             ld = d
             d += datetime.timedelta(days=14)
+        return ld
+
+    def get_cover_url(self):
+        ld = self._get_mag_date()
         url = ld.strftime(
             "https://www.sciencenews.org/wp-content/uploads/%Y/%m/%m%d%y_cover.jpg"
         )